1 / 32

Development of the Updated SAPRC Chemical Mechanism

This document provides an overview of the SAPRC mechanisms and their updates, including changes in ozone reactivity values and recommendations for improvement. It also discusses the major components of the SAPRC mechanisms and the mechanism generation system used. The objectives of the mechanism update are outlined, along with the updates made to the base mechanism, mechanism generation system, and aromatic mechanisms. This document is a comprehensive resource for researchers and airshed modelers.

losmond
Download Presentation

Development of the Updated SAPRC Chemical Mechanism

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Development of the UpdatedSAPRC Chemical Mechanism Outline • Overview of the SAPRC mechanisms • Updates for SAPRC-07 • Anticipated changes in ambient ozone predictions • Changes in ozone reactivity (MIR) values • Recommendations William P. L. Carter CE-CERT, University of California, Riverside, CA March 25, 2009 SAPRC-07 Mechanism Development

  2. Mechanism Use and Requirements Used in airshed models to: • Represent the chemistry of formation of secondary pollutants (e.g., O3) from emitted VOCs and NOx • Predict relative ozone reactivity scales for VOCs (D O3 / D VOC) for regulatory applications (e.g, MIR scale used in California) Requirements • Represent current state of science of atmospheric chemistry • Give predictions consistent with environmental chamber results • Appropriately represent the hundreds of types of VOCs for reactivity scale calculations • Represent complex ambient mixtures in airshed calculations in computationally efficient manner SAPRC-07 Mechanism Development

  3. Major Components of SAPRC Mechanisms Base Mechanism • Reactions of inorganics and common organic products Mechanisms for individual VOCs (hundreds of VOCs represented) • Mechanism generation system used to derive fully explicit mechanisms that are condensed using “lumping rules” • Estimated mechanisms are used for compounds whose explicit mechanisms cannot be estimated (e.g. aromatics) Condensed mechanisms for airshed models • Mechanisms for limited number of “lumped model species” derived from those of the mixture of individual VOCs they represent • These depend on the composition of the mixture taken as representative of VOC emissions from all sources SAPRC-07 Mechanism Development

  4. SAPRC Mechanism Generation System • Generates mechanisms for reactions of a wide variety of non-aromatic organics with OH, O3, NO3, O3P and by photolysis • Measured rate constants (or rate constant ratios) are used where available. Estimation methods are used otherwise. • Estimation Methods • Structure-reactivity estimates for OH, O3, and NO3 reactions • Nitrate yields from RO2+NO based on available (limited) data • Alkoxy radical estimates based (primarily) on correlations between rate constants and heats of reaction. • Other estimates based on extrapolations from known reactions or adjustments to fit chamber data • Limitations • Not applicable or reliable for all types of compounds • Currently only estimates reactions in presence of NOx SAPRC-07 Mechanism Development

  5. Major Objectives of Mechanism Update • Update rate constants and reactions to current state of science • Add chlorine chemistry • Improve mechanisms for aromatics to incorporate new data and improve performance in simulating available chamber data. • Update and enhance mechanism generation system were possible • Improve capability to be adapted to secondary PM models • Represent additional types of VOCs of interest to the CARB • Develop new condensed mechanisms from detailed version SAPRC-07 Mechanism Development

  6. Mechanism Updates --Base Mechanism • Inorganic and organic product mechanisms updated based on recent evaluations and data • Added Inorganic ClOx reactions and new model species to represent reactive Cl-substituted carbonyl oxidation products * Solar light SAPRC-07 Mechanism Development

  7. Mechanism Updates --Mechanism Generation System • Assigned VOC + OH, O3, NO3, and O3P rate constants updated • Capability to generate chlorine atom + VOC reactions added • Structure-reactivity methods to estimate rate constants • Mechanisms can be generated for saturated compounds, but alkene mechanism generation limited by lack of thermochemical group data for Cl-containing radicals • Capability to represent compounds and radicals with more than one ring and more than one double bond added • Used as aid in development of mechanisms for aromatics • Methods developed to generate initial reactions of OH, O3, and NO3 with the unsaturated dicarbonyl ring opening products • No methods exist to estimate unsaturated dicarbonyl photolyses, which need to be specified manually SAPRC-07 Mechanism Development

  8. Mechanism Updates --Aromatic Mechanisms • SAPRC-99 uses parameterized species with adjusted yields and photolysis rates for uncharacterized ring opening products. • Adjusted to fit aromatic - NOx experiments. • Model gives poor fits to effects of adding CO to aromatic - NOx experiments and to measurements of direct reactivity • SAPRC-07 aromatic mechanism update approach: • Estimate explicit mechanisms for ring-opening reactions and reactions of unsaturated dicarbonyl ring-opening products, based on data in Calvert et al. review to extent possible. • Condensed mechanisms for airshed models derived based on explicit mechanisms so developed • Adjust relative yields of photoreactive and non-photoreactive products based on simulations of chamber data SAPRC-07 Mechanism Development

  9. Ring Opening Mechanism Assumed Note: • Branching ratios do not depend on NOx levels • Yields of a-dicarbonyls basedon measurements • Unsaturated 1,4-dicarbonyl yields based on a-dicarbonyl co-product yields • Rest of ring opening is OH +diunsaturated dicarbonyl, formed with no NO to NO2 conversions SAPRC-07 Mechanism Development

  10. Alternative Ring Fragmentation Mechanism This mechanism predicts that the yields of a-dicarbolyls and the highly photoreactive unsaturated dicarbonyl products will depend on NOx levels. This is not consistent with product yield or chamber data. SAPRC-07 Mechanism Development

  11. Reactions of Aromatic Products • Parameterized mechanisms for phenols and cresols unchanged • Mechanisms for -dicarbonyls updated based on evaluations • Mechanism for model species representing unsaturated dicarbonyls derived based on estimated mechanisms for such compounds formed from the alkylbenzenes • For unsaturated 1,4-dialdehydes and aldehyde-ketones: • Assumed to photolyze rapidly based on data in Calvert et al. • Two model species used: one (AFG1) that photolyzes to form radicals and one (AFG2) that forms stable molecules • Relative yields of these adjusted based on simulations of chamber data for individual aromatics • For di-unsaturated 1,6-dicarbonyls and also 1,4-diketones: • Lumped into separate model species (AFG3) that is assumed not to photolyze, based on data in Calvert et al. SAPRC-07 Mechanism Development

  12. Mechanism Updates -- Improved Capability for Representing Low-NOx Products • Approximate methods are required to represent the peroxy + peroxy reactions important under low NOx conditions for the mechanisms to have reasonable size. • Because of the way this is represented in SAPRC-99 (like CB4/05), it is difficult to “un-lump” model species for organic hydroperoxides or organic nitrate products. • But higher molecular weight organic hydroperoxides and organic nitrates may be important precursors to SOA • SAPRC-07 uses a new representation of peroxy radical reactions that permits use of separate hydroperoxide and organic nitrate species, depending on volatility or other factors • For example, the current mechanism uses a separate model species for aromatic hydroperoxides, which may be useful to model NOx-dependence of SOA yields from aromatics SAPRC-07 Mechanism Development

  13. Mechanism Updates – Mechanisms for Many Types of VOCs Added or Improved • Number of distinct VOC mechanisms increased from 585 to 739 (a 27% increase) • Added or improved mechanisms include: • Added alkanes and alkenes in emissions inventories not previously represented • Separate estimated mechanisms for higher alkylbenzenes, depending on structure (SAPRC-99 used “lumped molecule” representation based on methylbenzenes) • Improved estimates for halogenated compounds, based on chlorine chemistry (also added CH3I) SAPRC-07 Mechanism Development

  14. Evaluation Against Chamber Experiments • Mechanism evaluated against available UCR and TVA environmental chamber data base + some UNC experiments • Performance simulating chamber data comparable to SAPRC99 (somewhat better in some cases, not quite as good in others) SAPRC-07 Mechanism Development

  15. Evaluation Results:Model Errors for All Single VOC Experiments SAPRC-07 Mechanism Development

  16. Evaluation Results:Model Errors for m-Xylene Experiments Average model error vs. hour of run SAPRC-07 Mechanism Development

  17. Evaluation Results:Model Errors for All Mixture Experiments SAPRC-07 Mechanism Development

  18. Model Performance SimulatingUCR-EPA Chamber Surrogate - NOx Runs • Problem of O3 underprediction at low ROG/NOx still exists SAPRC-07 Mechanism Development

  19. Lumped Mechanism for Airshed Models • A “Fixed Parameter” mechanism for airshed models was derived using same lumping approach as “Fixed Parameter” SAPRC-99 • Parameters for lumped species based on the base ROG mixture used in the reactivity scenarios • Mixture based on ambient air measurements in mid-1980’s (Probably needs to be updated) • Same mixture as used for “Fixed Parameter” SAPRC-99 • Emissions assignments have been updated and added to emissions speciation database • Mechanism slightly larger than SAPRC-99. • Used as base case in reactivity calculations SAPRC-07 Mechanism Development

  20. Effect of Update on Calculated Ozonein the Box Model Reactivity Scenarios SAPRC-07 Mechanism Development

  21. Condensed SAPRC-07 • Condensed versions of SAPRC-07 have been developed • Up to about half the number of species and reactions as uncondensed SAPRC-07. Comparable in size to CB05. • Essentially same O3 predictions for wide range of conditions. • Changes small for PANs, OH radicals, most other species. • Effects on H2O2 usually less than 10%, but depends on which peroxy radical lumping method is used. • Chemical validity is directly traceable to detailed SAPRC-07 • Based on detailed mechanisms for hundreds of VOCs • Can be readily modified if the base ROG mixture used to derive the lumped species parameters is changed • Can be readily updated if the underlying detailed mechanism is updated, by re-applying the condensation procedures. SAPRC-07 Mechanism Development

  22. Condensation Approaches Employed Two alternative peroxy radical representations: Other Condensations • Use representation in SAPRC-99 (and CB05) (CS07A) • Retain SAPRC-07 peroxy representation (CS07B) • Lump low reactivity products and other minor condensations • Lump Higher PANs with PAN2 (but PAN still explicit) • Use 1-product isoprene mechanism • Simplify aromatics mechanisms • Simplify chlorine mechanism • Reduce number of lumped alkane species for ambient mixtures SAPRC-07 Mechanism Development

  23. Results for Reactivity Scenarios SAPRC-07 Mechanism Development

  24. Calculation of Updated Reactivity Scales • Updated mechanism used to calculate MIR, MOIR, EBIR, and “Base Case” reactivity scales. • The scenarios and calculation methodology used is exactly the same as used for SAPRC-99 reactivity scales. • Reactivities calculated for a total of 1135 Types of VOCs, including • 739 VOCs represented explicitly • 320 VOCs represented by the “Lumped molecule” method • 76 Complex mixtures (e.g., exhausts, petroleum distillates) • This includes 358 types of VOCs not in 2002 SAPRC-99 scales. • New compounds of interest or or found in inventories • 24 Complex hydrocarbon mixtures derived to represent the 24 hydrocarbon “bins” used in CARB regulations SAPRC-07 Mechanism Development

  25. Changes in MIR Ozone Reactivity Values Base ROG Mixture: -5% Change SAPRC-07 Mechanism Development

  26. VOCs with Large MIR Changes * grams O3 / gram VOC SAPRC-07 Mechanism Development

  27. VOCs with Large MIR Changes(continued) * grams O3 / gram VOC SAPRC-07 Mechanism Development

  28. Recommended research • Base Mechanism • Reasons for model performance problems in simulating existing chamber data need to be elucidated • Uncertainty in OH + NO2 rate constant needs to be reduced • Uncertainty analysis and process analysis may be useful, but new approaches are needed. • Aromatics Mechanisms • Although improved, aromatics mechanisms are still highly simplified and not consistent with all current laboratory data • Improvements will probably require breakthroughs in both measurement methods and theory • Systematic examination of effects of alternative mechanisms may be useful to assess effects of uncertainties on models SAPRC-07 Mechanism Development

  29. Recommended research (continued) • Mechanism Generation System • The many estimation methods need to be updated, expanded and documented in the peer-reviewed literature • Predictions should be compared with those in the MCM • Methods to estimate volatility needs to be implemented to serve as a basis for lumping for SOA models • Adaptation for PM Modeling • Models for SOA formation need to be improved • Model species need to be added to represent low volatility SOA precursors • Well characterized chamber experiments with PM data at ambient concentration levels needed to evaluate mechanism SAPRC-07 Mechanism Development

  30. Recommended research (concluded) • Environmental Chamber Database • Need to take advantage of unique capabilities of UCR EPA chamber for PM and SOA Mechanism evaluation • Methods needed to improve sensitivity of chamber experiments to effects of oxidation products on reactivity • Mechanisms still uncertain for some types of compounds • Reactivity Scale Calculation • The work of the RRWG towards developing improved reactivity assessment methods needs to be completed. • At a minimum, the scenarios used for MIR calculations are way out of date and need to be updated. • The base ROG mixture is also out-of-date and needs to be updated SAPRC-07 Mechanism Development

  31. Related Anticipated CARB Research • Base Mechanism • Ongoing CARB contract could cover development of version of SAPRC-07 with lower k(OH + NO2) for sensitivity studies • Aromatic Mechanisms • Ongoing CARB Contract 07-730 calls for some work on developing improved aromatic mechanisms • Mechanism Generation System • No work in this area is presently funded • Adaptation for PM Modeling • CARB recently approved a project to adapt SAPRC-07 for PM modeling (David Cocker, co-investigator) SAPRC-07 Mechanism Development

  32. Related Anticipated CARB Research (continued) • Environmental Chamber Database • Project to adapt SAPRC-07 for PM modeling includes extensive set of SOA experiments in UCR EPA chamber • Methods to improve chamber reactivity experiments are being developed for CARB Coatings reactivity project • CARB Coatings project will cover reactivity experiments for additional classes of compounds • Reactivity Scale Calculation • The CARB is funding a project to update the base ROG • The EPA may be funding improving the box model scenarios used t calculate MIR • Work on improving the calculation methodology is presently unfunded SAPRC-07 Mechanism Development

More Related